Suicidal empathy

Why did the phrase spread?

Elon Musk and Bill Ackman gave the phrase reach. Musk spoke of “civilizational suicidal empathy” in a 2025 interview with Joe Rogan, and TIME notes he discussed “shallow vs deep empathy” with Tucker Carlson in October 2024 [1][3]; sources differ on the date of that interview. The book debuted at No. 1 on the New York Times Combined Print & E-Book Nonfiction list (31 May 2026) [1]. Wikipedia and TIME document the phrase as a term in public use; they are not scientific support for it.

Is it a scientific or clinical term?

No. According to TIME and UnHerd, Saad deliberately sets aside the research literature on empathy, calling its distinctions “unnecessary for our current treatment” [3][5]. The Global Project Against Hate and Extremism reports that the term is absent from peer-reviewed literature and that Saad has not run studies on it [4]; that organisation is an advocacy group with its own stance. An UnHerd review calls the book unscientific — vague definitions, anecdotes and films in place of data, “vibes-based” evidence [5]; that is one editor’s opinion, not a study.

What research says about empathy and bad decisions

The phrase has no research behind it, but the question under it — when does empathy lead to worse decisions? — does.

People give more to one identifiable victim than to statistical ones, and teaching them about this gap reduced giving to the identifiable victim without raising giving to statistical ones [6]; these were field experiments with students and small sums, and they say nothing about politics.

Across 41 studies the identifiable-victim effect is real but small (r = .05), stronger for a single victim, a child, or a photo [7] — empathy leans toward vivid cases rather than locking onto them.

Feeling and willingness to help peak for one child and drop from the second life on (“compassion fade”) [8]; this comes from lab scenarios, and it disappears when the group is presented as one unit.

Empathy for an out-group is muted compared with one’s own, and in competition can turn into pleasure at their misfortune [9]; this is a review, and the direction of the gap depends on context.

In the lab, people who induced empathy for a child moved her up a treatment queue, against the rules [10] — a hypothetical decision by students.

Researchers also discuss “pathological altruism”: “altruism in which attempts to promote the welfare of others instead result in unanticipated harm” [11]; it is a survey concept across 31 varied chapters, not a single measured construct.

Psychologist Paul Bloom argues that empathy works like a narrow, biased spotlight and is a poor guide for decisions [12]; it is an argument built on others’ data, and Bloom is not against compassion.

How is the phrase used in politics?

Critics read the phrase as political rhetoric: The Conversation and Salon place it among campaigns against empathy on the American right and describe it as an attack on a virtue [13][14] — commentary, not research. The main use Saad gives it, arguing about migrants and minorities, is a political claim, not a psychological finding [1][3]. Neither the book nor the research above shows that any group of people cares in a disordered way.

A better question to ask

A more useful question than “is empathy suicidal?” is one research can test: when does feeling for one vivid person pull a decision away from fair treatment of others? Bloom’s answer is “rational compassion” — keep caring, but check the decision against those outside the spotlight [12]; the queue experiment shows what that gap looks like in a lab [10]. Neither settles a policy question.